1. Field of the Invention
A method and system for testing reliability attributes in disk drives is disclosed. More specifically, the invention is a method and system for testing reliability attributes in disk drives based on margin magnitudes measured in a subset of the disk drives, the margin magnitudes indicating reliability for the population, or lot, of the disk drives.
2. Description of the Prior Art and Related Information
Companies in the field of hard disk drive manufacturing have recently seen an increase in pressure to lower costs. This pressure may be the result of lower profit margins per megabyte of storage, and increased competition in the market place.
As profit margins lower, manufacturers must increase manufacturing volume. With the increase in volume, there is at least a corresponding increase in costs in the factory where the hard disk drives are assembled.
Automated processes have helped to lower costs. For example, one process comprises writing a servo pattern on the magnetic media of the disk drive. Most manufacturers have employed large capacity servo formatters which can simultaneously connect to multiple disk drive assemblies to write the critical servo burst information that is used to position the read-write head of the a disk drive during operation.
A test apparatus is employed which is simultaneously connectable to multiple disk drive assemblies using multiple disk drive interfaces so that calibration, or tuning, processes and testing may be performed on the multiple disk drives at the same time. Such tests include tuning tests, optimization tests, and defect detection tests. The disk drives are formatted and may have customer specific product overlays installed while connected to the test apparatus.
Many of the tests that are performed on the disk drives are relatively time consuming. For example, some tests comprise on-going reliability tests (ORTs) which may comprise, complete multi-corner environmental stress tests used to ensure that the disk drives will operate reliably beyond the warranty life. Basic environmental stress tests may take 72 hours to complete. Another test includes a corrosion test suite which tests various corrosive properties of a disk drive by altering the environment in a chamber in which the disk drive is tested.
A system for testing reliability attributes in disk drives is disclosed. The reliability for the disk drives may be characterized as reliability attributes comprising measurements for predicting future reliability of the disk drives at the time of manufacture. The disk drives have reliability attributes and manufacturing attributes. The manufacturing attributes comprise process or component manufacturing attributes, which contribute to the reliability attributes of the disk drives.
The system comprises a servo formatter for preparing each disk drive with a servo format. A test apparatus is provided, to which each disk drive is connectable. The test apparatus includes one or more processors, and interfaces for connecting each disk drive to the test apparatus. Each of the disk drives comprises one or more processors which are capable of executing test instructions downloaded to the disk drives through the interfaces.
The test apparatus has a calibrator for tuning each disk drive connected to the test apparatus for verifying that each disk drive performs within expected limits. The calibrator may comprise software that executes on the processor of the test apparatus, and software that is downloaded to the disk drives for execution on processors of the disk drives. The tuning of the disk drives may include optimizing the disk drives for specific end uses. For example, tuning may comprise characterizing the gain component and frequency response of the read/write head of each disk drive. The electronics of each disk drive are thus tuned to conform with the individual head-media interface characteristics which vary from drive to drive.
A test containment logic may further be included with the calibrator for performing, or causing to be performed, one or more technical screen tests on each disk drive.
A work flow system is included, such as an integrated test system described in application Ser. No. 08/873,230 entitled xe2x80x9cMETHOD FOR MANUFACTURING A DISK DRIVExe2x80x9d filed Apr. 3, 2000, the full disclosure of which is incorporated by reference herein. The work flow system is connected to a network that is connected to the test apparatus through a network interface. The network may be connected to a plurality of test apparatuses which communicate with the work flow system through the network. A randomizer for automatically selecting a first sample set of the disk drives is provided with the work flow system. The randomizer may comprise a software module that executes on a processor of the work flow system that randomly selects the disk drives to be included in the first sample set. The first sample set has a size programmed into the randomizer for defining the size of the first sample set, which may be distributed across more than one test apparatus. The work flow system communicates with the test apparatus to define which disk drives connected to the test apparatus should be included in the first sample set.
An extended testing logic is included with the test apparatus for executing one or more extended tests on the first sample set of disk drives for measuring a margin magnitude for one or more of the reliability attributes. The extended testing logic comprises a software module for execution on the processors of the selected disk drives in the sample set, and may include some instructions which execute on the processor of the test apparatus. The margin magnitude is used as a predictor of reliability for the disk drives wherein a higher performance level suggests a high reliability and a low performance level suggests a low reliability. For example, a disk drive having a high mean number of seek operations between errors may tend to have a high reliability.
The extended testing logic retrieves a reference for the margin magnitude and detects a significant change in the margin magnitude from the reference. The reference may be calculated by, for example, averaging the margin magnitude for each extended test performed over time. The reference may alternatively be determined by selecting and specifying a minimum performance parameter based on general manufacturing experience among engineers for predicting hard disk drive reliability.
When compared with the reference, if a significant change is detected in the margin magnitude from the reference by one or more of the extended tests, the extended testing logic sends a request to the work flow system to automatically adjust the defined sample size to use for the size of a second sample set of the disk drives. The randomizer is further adapted to automatically select the second sample set to be communicated to the test apparatuses. The second sample set has the newly adjusted defined size. The extended testing logic is further adapted to repeat the extended tests on the second sample set to validate the significant change that was detected with respect to the first sample set.